The Application of Multiobjective Evolutionary Algorithms to an Educational Computational Model of Science Information Processing: a Computational Experiment in Science Education

Conflicting explanations and unrelated information in science classrooms increase cognitive load and decrease efficiency in learning. This reduced efficiency ultimately limits one’s ability to solve reasoning problems in the science. In reasoning, it is the ability of students to sift through and identify critical pieces of information that is of paramount importance in science and learning. Unfortunately, the ability to accomplish the identification of critical ideas is not one that develops without practice and assistance form teachers or tutors in the classroom. The purpose of this paper is to examine how the application of an evolutionary algorithm works within a cognitive computational model to solve problems in the science classroom and simulate human reasoning for research purposes. The research question is: does the combination of optimization algorithms and cognitive computational algorithms successfully mimic biological teaching and learning systems in the science classroom? Within this computational study, the author outlines and simulates the effects of teaching and learning on the ability of a “virtual” student to solve a science task. Using the STAC-M computational model the author completes a computational experiment that examines the role of cognitive retraining on student learning. The author also discusses the important limitations of this powerful new tool.